V53B-1319
Shear-Wave Splitting From Local Earthquakes as an Indicator of Crustal Stress at Yellowstone
Our analysis of split shear waves from local earthquakes on the Yellowstone Plateau is consistent with the complex stress and strain fields determined from previous studies. Splitting fast directions (φ) and delay times were determined using a cross-correlation algorithm for a subset of data from well-located earthquakes recorded from 1993-2006. Seismic anisotropy in the crust can be caused by aligned platy minerals, layering of bedding or foliation, and aligned microcracks. Stress in the crust aligns microcracks such that cracks with faces normal to the maximum compressive stress (σ1) close, while cracks with faces oriented normal to the minimum compressive stress (σ3) may open. This creates a stress-dependent anisotropy where φ is perpendicular to σ3. Previous studies, which used earthquake focal mechanism inversion, GPS, and InSAR, showed that crustal stresses and strains at Yellowstone are influenced by regional extension as well as transient deformation. The σ1 direction is generally near vertical, but σ3, which is nearly horizontal, varies from roughly N-S near the surface rupture of the 1959 M7.5 Hebgen Lake earthquake to NE-SW near the rim of the Yellowstone Caldera 35 km east of the fault. The agreement between our shear-wave splitting results and previous stress and strain field studies suggests that the inferred anisotropy is due to stress-oriented microcracks in the upper crust. The spatial variation in the σ3 direction has been modeled as due to postseismic viscoelastic relaxation in the lower crust-upper mantle following the Hebgen Lake normal-faulting earthquake combined with regional NE-SW extension. Changes in the directions of anisotropy observed at Ruapehu Volcano in New Zealand over a few years were attributed to changes in the stress field associated with pressure changes in the magmatic system. But despite well-documented temporal deformation on similar time scales at Yellowstone there are no corresponding temporal trends in the shear-wave splitting data. As the number of three-component stations in Yellowstone increases, we will be more likely to detect possible temporal changes in anisotropy.
V53B-1320
Measurement of crustal flexure in the Lake Hills, South Central Idaho and Timing of Eastern Snake River Plain Subsidence
Late Miocene rhyolite along the north flank of the Eastern Snake River Plain (ESRP) near Carey, Idaho, was studied to investigate the flexure and subsidence history of the ESRP. The rhyolite has been divided into three formations that include seven individual ignimbrite flows. A petrology study of 26 vitrophyre samples, from the lower 4 flows, revealed 3 distinct formations based on percentage of small, large and skeletal feldspars, degree of welding and percentage of glomerocysts. Formations Tiv O (oldest) and Tiv Y (youngest) are indistinguishable in hand sample and in outcrop; however, the middle formation, Tiv M, harbors a distinct phenocryst-poor and thin vitrophyre. New 40Ar/39Ar analysis yields ages of 9.21±0.18 Ma and 9.16±0.20 Ma for Tiv O, 8.39±0.54 Ma for Tiv M and 8.76±0.38 Ma for Tiv Y. Field mapping and measurement of ignimbrite compaction foliations delineates three structural domains. Together the three domains exhibit a map-scale east-trending flexural antiform. Domain 1 extends 0-4 km north of the ESRP. Stratigraphically up section, southerly dips have an average of Tiv O 10°, Tiv M 7°, Tiv Y 7° and for Quaternary basalt 5°. With a slightly undulating topography, Domain 2 stretches 4-6 km north of the ESRP and displays dips of 10-21° to the north and south. All unit groups are present within Domain 2; however, in numerous locals Tiv M is directly underlain by Challis volcanics (Tcv) and where Tiv O is present, the unit is thinner. Domain 3 extends 6-12km north of the ESRP and has predominantly north dipping foliations between 10-21°. As in Domain 2, Tiv O is thinner or not present throughout domain 3. We interpret the formation of the antiform as a crustal flexure response to the subsidence of the ESRP. The northward thinning of Tiv O, suggests a topographic high in the previous paleotopography within Domain 2 and the initiation of the subsidence of the ESRP prior to rhyolite deposition. Two distinct angular unconformities between Tiv O-Tiv Y and Tiv Y-Qb, further indicate that subsidence continued during and after the emplacement of the rhyolite. Previous studies of crustal flexure along the northern edge of the ESRP have been conducted at Howe Point (100 km NE of the Lake Hills) and Lidy Hot Springs (130 km NE of Lake Hills). Together with the new data from the Lake Hills presented within, over 130 km of crustal flexure has been identified and analyzed along the northern boundary of the ESRP. The Lake Hills experienced flexure before, during and after ignimbrite deposition. Two episodes of flexure occurred at Howe Point. Major flexure (25°) occurred from 16-10 Ma while minor flexure commenced after 6.0 Ma. Lidy Hot springs experienced major flexure before 7-10 Ma and minor flexure after 6.0 Ma. Major flexure and ESRP subsidence has been was previously proposed to signify an isostatic crustal response to the emplacement of plutonic loads below the ESRP. However, initiation of crustal flexure preceded the emplacement of Yellowstone hotspot ignimbrites in all three locations. Only minor flexure has been identified in all three locations during or after ignimbrite deposition, suggesting plutonic loading had already occurred prior to ignimbrite deposition.
V53B-1321
New 40Ar/39Ar age Determinations of Basalts From the Yellowstone Plateau Volcanic Field
The Yellowstone Plateau volcanic field is one of the largest and most active silicic volcanic systems in the world and is the youngest manifestation of the Snake River Plain-Yellowstone hotspot track of bimodal basaltic-rhyolitic volcanism. Yellowstone's volcanic history is marked by three cataclysmic caldera-forming eruptions at 2.05, 1.29, and 0.64 Ma. Each volcanic cycle has been interpreted to result from the accumulation, storage, and differentiation of discrete silicic magma reservoirs. The latest cycle, forming the 2900-km2 Yellowstone caldera was followed by at least 30 large rhyolite flows as young as 70,000 years old [Christiansen, 2001]. Coeval basalts are exposed around the margins of the Yellowstone Plateau, mainly north of Yellowstone caldera and in the Island Park area southwest of Yellowstone caldera. The total estimated volume of basalts in the volcanic field is about 250 km3. The basalts of the field are generally low-potassium olivine tholeiites; most have plagioclase and olivine phenocrysts and all contain olivine, clinopyroxene, plagioclase, and iron-titanium oxides in the groundmass. Basalt chemistry ranges from 46 to 50 wt% SiO2 and 0.3 to 0.5 wt% K2O. A more quantitative understanding of the relation between basaltic volcanism and Yellowstone's rhyolitic cycles is hampered by a lack of high-precision geochronology. Previous age determinations of Yellowstone basalts were based on stratigraphy, paleomagnetic polarity, and K/Ar dating, and two 40Ar/39Ar ages of basalt flows coeval with the third cycle (Smith and Bennett, 2006). We present 12 new 40Ar/39Ar ages representing seven different basalt units (Christiansen, 2001). Samples were selected based on their lack of alteration, groundmass texture, and potassium concentrations. Age determinations were made on separated crystalline groundmass and are a mixture of plateau and isochron ages. The new 40Ar/39Ar ages (1 σ errors) are 835±17 ka for basalt of Warm River, 650±9 ka for Undine Falls basalt, 743±5 ka and 725±25 ka for Snake River basalt, 380±30 ka for Madison River basalt, 260±12 ka for Osprey basalt, 254±12 ka to 108±15 ka (n=4) for Swan Lake Flat Basalt, and 151±7 ka for Gerrit basalt. These new age determinations provide more rigorous constraints on Yellowstone's volcanic history, and can be incorporated into assessments of Yellowstone's volcanic hazards. Christiansen, U.S. Geol. Surv. Prof. Pap. 729-G, 2001, 145 p. Smith and Bennett, Yellowstone Science, 2006, 14, 5-12.
V53B-1322
Individual magma batches related to the origin of Yellowstone 516-70 ka post-caldera rhyolitic lava flows : results from glass geochemistry.
Since its last collapse at 640 ka, Yellowstone caldera has been filled by at least 25 voluminous rhyolitic lava flows totalling more than 900 km3. The oldest identified flows have been erupted at 516 ka, 486-479 ka and 198 ka around the two resurgent domes. Most of the known effusions occurred during relatively short episodes from 165 to 147 ka, 117 to 102 ka, and 72 to 70 ka (Christiansen, 2001), erupting along 2 NNW-trending lineaments which are parallel to regional faults and do not intersect the resurgent domes. The intermittent nature of the volcanism and its spatial distribution remain largely unresolved questions. We used a combination of XRF whole rock major and trace element analyses, electron microprobe and laser ablation-ICP-MS major and trace element glass micro-analyses to examine the temporal and spatial links among the eruptions. Petrologically, we observe a broad-scale differentiation of the lavas over time. Lavas dated at 516 ka have SiO2 contents of 71-72 wt% for bulk rock and 76-76.5 wt% in glass, while 165 ka and younger lavas exhibit SiO2 contents of 76.5-77.5 wt% for both the bulk rock and glass. Mineralogically, the lavas evolve from plagioclase-dominated to sanidine and quartz with no plagioclase. These observations suggest an origin from a common source undergoing slow differentiation. Temporal changes of certain trace elements such as Sr and Ba also suggest a similar evolution. For instance, Sr evolves from 115 ppm in bulk rock in the most primitive lavas of the 516 ka group, to 60 ppm at 198 ka, declining to 2.5 ppm at 70 ka. A simultaneous increasingly negative Eu anomaly is observed in the glass. Such behavior is expected from feldspar fractionation in a long-lived reservoir. Elements such as Y, Th and REE also behave compatibly in the Yellowstone lavas, their concentrations being systematically higher in bulk rock compared to glass. Should the system be a long-lived magma chamber undergoing fractionation, their concentrations over time would therefore decrease. However, Y generally increases from 60 ppm in bulk rock at 516 ka to 65-70 ppm at 165-147 ka and 85 ppm at 70 ka. Of the analyzed elements, only Rb exhibits concentrations higher in glass than bulk rock and therefore is incompatible. Although it increases slightly over time in bulk rocks from 160 ppm in the most primitive of the 516 ka lavas to 170 ppm at 198 ka, it is generally buffered at 180-200 ppm from 165 to 70 ka. In addition, spatial compositional differences are observed between the flows, despite comparable mineralogy and crystal contents. For instance, during the 165-147 ka period, Ba in the western lavas ranges from 50 to 70 ppm, while it varies from 95 to 120 ppm in the eastern lavas. During the 117-102 ka period, two western lavas erupted at 6 km distance from each other at 112 and 108 ka have respective contents of 40 vs 80 ppm Ba in glass, 40 vs 160 ppm Ba in bulk rock and 65 vs 80 ppm La in glass. These spatial compositional differences, together with the decoupled temporal changes in various trace elements, suggest that these lava flows did not originate from a single homogeneous long-lived liquid reservoir. It is possible that eruptions are the result of localized magma injections at various depths into the mush, which mixed with the mush or merely provided heat to remelt the mush and initiate magma ascent. Each batch of magma may then have undergone variable degrees of differentiation before eruption.
V53B-1323
Storage of Explosive versus Effusive Rhyolite Magma at the Yellowstone Volcanic Center
The Yellowstone volcanic center has erupted more than 900 km3 of rhyolitic magma in the last 600,000 years (1). Most of that magma extruded as large lava flows, with only a few known explosive eruptions. Why have explosive eruptions been so rare in the recent history of the Yellowstone volcanic system? To explore that question, we focus on the Tuff of Bluff Point (TBP), about 50 km3 of magma that explosively erupted 173 ka, forming the West Thumb caldera (1). Like most other recent eruptions of Yellowstone, TBP is high silica rhyolite, with phenocrysts of quartz, sanidine, and minor ferro-pyroxenes and Fe-Ti oxides. Fe-Ti oxide and pyroxene compositions indicate that the magma had equilibrated at an oxygen fugacity equal to the QFM buffer. Rehomogenized glass inclusions (n=7) in quartz contain 2.2-3.1 wt.% water and between 400-650 ppm CO2. Those volatile contents indicate storage pressures of 90-160 MPa. Ubiquitous pyrrhotite shows that the magma was sulfur saturated, and most likely volatile saturated. The co-existing fluid would be only 42-47% water. Cathodoluminescence (CL) images of quartz phenocrysts reveal mainly concentric growth zones, with occasional dissolution boundaries present. Ti contents in quartz generally decrease from core to rim, indicating cooling of the magma, although the relative temperature changes recorded are only 10-15°, with only minor changes across dissolution boundaries. To put our observations in perspective of the recent Yellowstone magma system, we have begun examining some of the recent rhyolitic lavas, including the Pitchstone Plateau (PP), a single homogeneous lava flow of 70 km3 that erupted 79 ka (1). CL images also reveal mainly concentric quartz growth, with few dissolution boundaries obvious. Ti contents in quartz also generally decrease from core to rim, but are uniformly lower than in those in TBP, suggesting that PP magma was colder than TBP magma. Glass inclusions (n=20) in PP are generally water poor and rarely contain CO2. A few do have more than 2 wt.% water, but only100-200 ppm CO2, indicating storage pressure of 80-100 MPa. (1) Christiansen et al., USGS Open-file Report 2007-1071, 2007, 94 p.
V53B-1324
Olivine Crystallization and Mantle Potential Temperatures Beneath Yellowstone
New basalt samples from the Yellowstone Plateau volcanic field provide evidence for some of the most primitive liquids yet recovered for the region, and yield clues regarding mantle processes. The sample distribution covers a large area and an extended period, and one sample in particular (basalt of Warm River) contains 11% MgO, with olivines that are in equilibrium with the host whole rock. Using olivine thermometry, we calculate both olivine crystallization and mantle potential temperatures (Tp, the temperature a parcel of mantle would have if it rose adiabatically to Earth's surface without melting) to test whether the alleged Yellowstone hot spot is truly hot. These tests make use of thermometers from (1) and (2), and we compare temperatures at Yellowstone with estimates from the Hawaii Scientific Drilling Project, HSDP-2 (2, 3) and the Siqueiros Transform, near the East Pacific Rise (4). Assessment of olivine-liquid equilibrium is based on the Fe-Mg exchange coefficient between olivine and liquid, which is assumed to be 0.30+/-0.03 (5). In total, the Yellowstone lavas have mean crystallization temperatures of 1251+/-41oC (n=79) with a maximum of 1327oC. The mean temperature is similar to crystallization temperatures of basalts from Siqueiros (1264+/-21oC), but lower than the mean temperature for HDSP samples (1343+/-50oC). Mantle potential temperatures appear to approach an olivine-control line, which if valid, yields a mantle potential temperature of 1610oC, slightly higher than most Snake River Plain (SRP) lavas (Tp =1540oC). Applying the same model to lavas from the Siqueiros Transform yields a Tp of 1400oC, and so excess temperatures (relative to MORB) along the SRP are in the range of 140-209oC, consistent with a mantle plume interpretation for the Yellowstone hot spot track. These calculations presume that primitive melts have equilibrated with mantle olivine of Fo90 in composition; given the FeO contents of SRP lavas, parental liquids should have between 15-18% MgO. We tentatively conclude that the Yellowstone hot spot is indeed hot compared to ambient mantle. (1) Beattie (1993) CMP, 115, 103-111. (2) Putirka et al. (2007) Chem. Geol., 241, 177-206. (3) Rhodes and Vollinger (2004) G-cubed, 5, 10.1029/2002GC00434. (4) Perfit et al. (1996) EPSL, 141, 91-108. (5) Roeder and Emslie (1970) CMP, 29, 275-289.
V53B-1325
A GIS Framework for Mitigating Volcanic and Hydrothermal Hazards at Yellowstone National Park and Vicinity
Yellowstone National Park encompasses one of Earth's largest systems of volcanic, seismic, and hydrothermal activity. These active hydrothermal and volcanic systems, and associated seismicity have the possibility for future violent events and pose potential hazards to park visitors and infrastructure. Depending on the nature and magnitude of a particular hazardous event, from small-localized hydrothermal steam explosions to very large volcanic eruptions of ash and lava, and the particular time and season when it might occur, 70,000 to more than 100,000 people could be affected. Although the most violent events are the least likely, their occurrence could affect a broader region or even continent-wide areas. The Yellowstone Volcano Observatory (YVO) was created to monitor these hazards. A Geographic Information System (GIS) is one of the tools YVO is using to assess and monitor hazards. A GIS provides a central storage location for data such as: location and distribution of lava flows, ash deposits, and pyroclastic flows, seismicity, Global Positioning Satellite (GPS) and other geodetic data, and instrument locations, legacy surveys, and park infrastructure. The system also provides tools for the creation of deformation surfaces, hazard buffers, hazard modeling, and portrayal of complex hazard information to local officials and the public. Continued monitoring by YVO will facilitate recognition of premonitory indications before a volcanic eruption. GIS will play a role in both the monitoring and mitigation of the hazards in Yellowstone National Park and vicinity by facilitating advance preparation for responses to potential hazards from future volcanic eruptions or hydrothermal explosions and can help mitigate their effects on the park, its staff and visitors.
V53B-1326
A Prototype Hydrothermal Monitoring System, Norris Geyser Basin, Yellowstone National Park, Wyoming
Hydrothermal explosions are a prominent geologic hazard in Yellowstone National Park and are of consideration for park infrastructure and visitor safety. It is estimated that small rock-hurling phreatic explosions occur somewhere in the park almost every year and larger basin-wide events on the order of several hundred years. The Yellowstone Volcano Observatory (U.S. Geological Survey, University of Utah, and the National Park Service) has deployed a prototype network of GPS and seismic stations in Norris Geyser Basin. The monitoring system consists of five GPS stations and one broadband seismograph that were installed and operated for a year (October 2006 through September 2007) including during Yellowstone's harsh winter. The five GPS stations operated remarkably well over the survey period with at least 3 stations operating 98% of the time. The general southwest horizontal motion and subsidence of the 5 GPS stations are consistent with observations from nearby permanent GPS stations and InSAR. However, local transient signals of uplift and subsidence up to 6 cm are observed. Various long-period signals are observed in the seismic data, ranging from 2 to 100 seconds, which may be indicative of the transport of hydrothermal fluids within the basin. GPS and seismic data will be analyzed and compared to available temperature (air and water), rainfall, and barometric pressure data to try and isolate signals that can be attributed to the hydrothermal system. Ground deformation data can be used to determine the interdependence between regional deformation and hydrothermal activity. Seismic data can be used to help determine the interdependence between regional earthquakes and hydrothermal activity. These data will be valuable to YVO to help us better monitor Yellowstone's many hydrothermal systems to both gain a greater understanding of how they work as well as to be able to better understand the safety hazards involved to both park employees and visitors.
V53B-1327
Using Continuous Monitoring of Ambient CO2 and H2S to Assess Toxic Gas Hazards in Yellowstone National Park, Wyoming, USA
The mysterious death of five bison in the Norris Geyser Basin area of Yellowstone in 2004 was apparently due to an increase in the output of CO2 or H2S, coupled with unusually cold, still weather. This event of nature may support a long-held claim of geochemists: near-surface changes in pressure, temperature, hydrologic flow, ground permeability, and wind conditions can reasonably be expected to produce attendant variability in the output and ambient concentration of gases emitted from hydrothermal areas. Monitoring these changes at the surface provides a window to processes occurring below, and a continuous assessment of gas hazards for frequently visited places like Norris. To characterize subsurface processes and identify hazards, we developed a transportable monitoring system to measure ambient gas concentrations and meteorological parameters. The solar-powered system uses industrial grade sensors for CO2 and H2S gas, along with sensors for wind speed and direction, barometric pressure, and ambient temperature. In order to reduce power use and prolong sensor life, every 10-minutes the gas sensors are powered on and allowed to stabilize, and the average values for the gas and met sensors are then recorded. The system can be configured for on-site data logging or radio telemetry. During the first year of operation in a thermal area adjacent to where the bison died, the system recorded diurnal variations. Although CO2 build-up was observed at night during cool windless conditions, ambient concentrations of CO2 and H2S remained below hazardous levels. Encouraged by the robust performance of the sensors, a second system was built to use as a roving monitor within the park as conditions permit and opportunities arise to track thermal variations. The performance of this system during the first year of operation reinforces the importance of continuous monitoring for processes such as gas-release events. Such occurrences, while evidenced in nature by events like the bison deaths, can be both acute and fleeting, and might go unnoticed by campaign-style survey measurements. Economical, low impact, sturdy monitors like the one described here can supplement GPS, seismic and other continuous monitoring, and can help fill in temporal holes of synoptic measurements.
V53B-1328
Continuous real-time monitoring of chloride in geothermal areas in Yellowstone National Park: initial results from newly developed long-term in-situ chloride analyzers.
Chloride in the surface waters of Yellowstone National Park is primarily derived from magmatic/hydrothermal sources. Discrete chloride measurements, collected at weekly to monthly intervals, are a key component of the ongoing geothermal monitoring program conducted by Yellowstone Volcano Observatory and National Park Service scientists. Chloride flux, estimated from discrete chloride measurements and streamflow data, could potentially be used as a proxy for geothermal heat flux and volcanic-geothermal unrest in the Park. However, infrequent chloride sampling restricts our understanding of dynamic geothermal processes, and the lack of real- time chloride data limits our ability to provide early warning and timely response to geologic hazards in Yellowstone. We seek to combine real-time chloride and streamflow data to examine variations in chloride flux due to changes in the volcanic-geothermal system and to determine if real-time chloride flux data can be used as an early warning indicator of volcanic hazards in the park. To address these objectives, we have developed a low-cost instrument for long-term, real-time, in-situ chemical analysis, the Field Sequential Injection Analyzer (Field-SIA). The Field-SIA is self-calibrating, performs hourly analyses for over two months between service visits, and integrates with existing USGS streamflow gaging stations which provide solar power and satellite telemetry of real-time chloride data. The Field-SIA greatly increases chemical data collection while significantly decreasing the cost of sampling and analysis. We will present data from long-term, high-resolution, real-time chloride monitoring of: 1) Tantalus Creek which drains the Norris Geyser Basin; 2) the Firehole River which drains the Upper, Middle, and Lower Geyser Basins; and 3) the Yellowstone River near Gardiner, MT. Initial results suggest that chloride fluctuations at Tantalus Creek were linked to diel temperature cycling while chloride fluctuations at the Firehole and Yellowstone River sites were more closely related to river flow.
V53B-1329
Mapping Changes in Yellowstone's Geothermal Areas and Radiative Flux
Yellowstone National Park contains the world's largest concentration of geothermal features, with an estimated more than 10,000 features. The National Park Service is legally mandated to protect and monitor these natural features, and a geothermal monitoring plan including remote sensing has been approved. Inexpensive, accurate, and efficient geothermal mapping and change detection techniques are being developed to aid in monitoring geothermal features. Geothermal features are constantly changing in size, shape, distribution, and radiative flux. We are examining the change in geothermal activity in Yellowstone National Park and surrounding areas from up to 30 years ago to the present. Possible drivers of change include seismic activity, climate, geothermal energy development outside the park, and proximity to the caldera boundary. We are mapping and documenting changes in geothermally active areas and geothermal radiative flux using Landsat Thematic Mapper (TM), Enhanced Thematic Mapper Plus (ETM+), and potentially Multispectral Scanner (MSS) satellite imagery. We are using change vector analysis to study change in recent years starting with 2007 as our base year, as well as historic change possibly as far back as 1979. These maps will allow us to evaluate hypothesized drivers of change in geothermally active areas by determining whether observed spatial patterns are consistent with patterns expected from these drivers.
V53B-1330
Volatile Emissions from Hot Spring Basin, Yellowstone National Park, USA
The flux and composition of magmatic volatiles were characterized for Hot Spring Basin (HSB), Yellowstone National Park, in August 2006. Diffuse fluxes of CO2 (228 sites) from thermal soil were elevated, with a population distribution similar to that of other acid-sulfate areas in Yellowstone. Thus the estimated diffuse emission rate at HSB is proportionately larger than other areas due to its large area, and could be as high as 1000 td-1 CO2. The diffuse flux of H2S was only above detection limits at 20 of the 31 sites measured. The estimated diffuse H2S emission rate was ~ 4 td-1. Good correlation exists between the log of CO2 flux and shallow soil temperatures, indicating linked steam and gas upflow in the subsurface. The correlation between CO2 and H2S fluxes is weak, and the CO2 / H2S diffuse flux ratio was higher than in fumarolic ratios of CO2 to H2S. This suggests that various reactions, e.g., native sulfur deposition, act to remove H2S from the original gas stream in the diffuse low- temperature environment. Dissolved sulfate flux through Shallow Creek, which drains part of HSB, was ~ 4 td-1. Comparing dissolved sulfate flux to estimates of primary emission of H2S based on fumarolic gas geochemistry gives first order estimates of the sulfur consumed in surficial or subsurface mineral deposition. Total C and S outputs from HSB are comparable to other active volcanic systems.
V53B-1331
Airborne CO2 and H2S Measurements at Hot Spring Basin, Yellowstone National Park
Gas emission-rate measurements at thermal areas located in remote regions with difficult ground access and little topographic relief pose a special challenge to those attempting to assess volcanic hazards in those areas. Several attempts have been made to measure gas emission rates from geyser basins, thermal areas and discrete large fumaroles at Yellowstone National Park through the use of fixed-wing aircraft with an on-board measurement system similar to that employed elsewhere at large stratovolcanoes. Despite minimum flight elevation restrictions and relatively flat terrain that often make access to the lowest margins of the plume difficult in these areas, we have successfully measured plumes of CO2 and H2S at several such areas and features at Yellowstone. We report here the results of a series of airborne measurements on 7 Jun 2006 at Hot Spring Basin (HSB), a remote vapor-dominated hydrothermal system just outside the northeast margin of Yellowstone caldera containing multiple gas vents. Using a LI-COR infrared spectrometer and Interscan electrochemical detector system, we detected a 3-km-wide plume approximately 2 km downwind from HSB. Several airborne traverses through a vertical slice of the plume allowed us to construct a cross-section of the plume and yielded emission rates of 170 metric tonnes per day (t/d) for CO2 and 2 t/d for H2S, similar to rates measured at Mt. Baker, WA (USA) in September 2000. However, an August 2006 ground-based study of emissions from HSB yielded higher emission rates for both CO2 and H2S (Werner et al., this session), suggesting that not all of the diffuse emissions are reflected in the airborne measurement. Although a complete inventory of plume emission rates from the majority of degassing sources in Yellowstone National Park is not yet complete, HSB appears to be a smaller gas emitter than some of the other sources in the Park (e.g., Norris Geyser Basin, Brimstone Basin, Mud Volcano, Grand Prismatic Spring and Mammoth Hot Springs).
V53B-1332
Twenty Five Years of Chloride Flux Data From the Yellowstone River Drainage: Temporal Trends and Spatial Distribution
Chloride flux is one component of Yellowstone National Park's Geothermal Monitoring Plan. It has been used as a measure of volcanic and geothermal activity by a variety of researchers. Chloride flux is a measure of the mass of chloride, predominately from the magmatic system, leaving a basin during a given time period and is determined by multiplying chloride concentrations from river samples by USGS stream discharges values. Yellowstone's four major drainages (Yellowstone, Madison, Falls and Snake) have been sampled approximately 25 times a year since 1983. The Yellowstone River drainage above Corwin Springs encompasses an area of 6,800 km2 and ranges in elevation from 1,550 to 3,400 meters. The andesitic Absaroka Range makes up the majority of the basin, but there is a significant area of Quaternary rhyolite associated with the Yellowstone Volcano. It is a snowmelt- dominated system with a mean annual flow of 88 m3 sec-1 (3,100 cfs). Yellowstone Lake (350 km2), accounts for five percent of the drainage area and is the dominant water source during base flow periods. High-chloride and acid-sulfate, thermal springs predominately are found near the Yellowstone Caldera boundary and within the Norris-Mammoth corridor. Within the Yellowstone River drainage, chloride flux has been measured at the outlet of Yellowstone Lake, Lamar River, Gardner River and Boiling River. Each of these sites has different chloride concentrations and different temporal trends. We have observed that instantaneous chloride flux increases with increased discharge in the Yellowstone River at Corwin Springs and at the Yellowstone Lake outlet. Years with greater total water yield correspond to years with greater total chloride flux. The chloride flux data show that the variation in chloride flux within one year is greater than the variation in total chloride between years. Synoptic sampling of Yellowstone River tributaries and transects of chloride concentrations along the Yellowstone give insight into the surface locations that produce the approximate 1.7 x 1010 grams of chloride that flow down the Yellowstone River each year.
V53B-1333
Development of a Wireless Network of Temperature Sensors for Yellowstone National Park (USA)
Temperature sensors deployed at Yellowstone clearly document that thermal features can vary in temperature on a variety of timescales and show regional correlations unrelated to meteorological variables such as air temperature. Yellowstone National Park (YNP) staff currently measures temperatures at over 40 thermal features and streams within the park, utilizing USGS stream gaging stations and portable data loggers deployed in geyser basins. The latter measure temperature every 1 to 15 minutes, and the data are physically downloaded after about 30 days. Installation of a wireless sensor network would: 1) save considerable time and effort in data retrieval, 2) minimize lost data due to equipment failure, and 3) provide a means to monitor thermal perturbations in near-real time. To meet this need, we developed a wireless sensor network capable of in-situ monitoring of air and water temperature. Temperature sensors are dispersed as nodes that communicate among themselves and through relays to a single base-station linked to the Internet. The small, weatherproof sensors operate unattended for over six months at temperatures as low as -40°C. Each uses an ultra-low-power Texas Instruments™ MSP430 microcontroller and an SD card as mass storage. They are powered by 15Ah, 3.6 v, inert Li-ion batteries and transmit data via 900MHz radio modules with a 1-km range. The initial prototype consists of 4 nodes, and is designed to scale with additional nodes for finer spatial resolution and broader coverage. Temperature measurements are asynchronous from node to node, with intervals as frequent as 30 seconds. Data are stored internally to withstand temporary communication failures; underlying intelligent software is capable of re-routing data through alternative nodes to the base station and a MySQL data archiving system. We also developed a Google-Maps-based, front-end that displays the data, recent trends and sensor locations. The system was tested in the Santa Cruz Mountains and will be used at Yellowstone National Park during Fall 2007.
V53B-1334
Castle Geyser and Bobby Sox Trees: Pulses and Pauses in the Development of Hydrothermal Features in the Upper Geyser Basin, Yellowstone National Park, Wyoming
Preliminary 14-C dating of Castle Geyser, combined with observations of living and dead trees in hydrothermal areas, suggests that hydrothermal systems in Yellowstone have pulses of activity interspersed with pauses of little or no activity. Between the time scale of volcanic activity, with pulses and pauses over thousands to hundreds of thousands of years, and geyser eruptions, with pulses and pauses over minutes to decades, lies the time scale for pulses and pauses in the development of individual hydrothermal systems and large thermal basins. Castle Geyser has long been noted as being among the largest, and therefore probably oldest, geysers in Yellowstone. Watson (1961) proposed an age of 8000 years for the geyser cone, and Bryan (2001) suggested that it is 5000 to 15000 years old. Recent dating, accompanied by 3-D laser mapping, suggests a complex, multi- stage development of the geyser. AMS 14C dating of microbial and pollen carbon trapped in siliceous sinter that forms a broad, gently-sloping shield at the base of the geyser cone yields ages of 8787 +/- 60 years BP and 10472 +/- 70 years BP. Carbon from sinter on the cone of the geyser yields ages equal to or younger than 1038 +/- 35 years BP. No samples dated so far have ages between 8787 and 1038 years BP. The morphology of the geyser suggests that the pause after shield formation was followed at least one stage of terrace formation (from either hot spring or pool-type geyser activity), which in turn has been followed by the construction and partial destruction of a massive cone. Where thermal waters are high in silica, thermally killed trees may develop white lower trunks, informally known as "bobby sox." Forest growth implies a time of no thermal activity; forest death, where clear evidence of thermal activity exists, implies inception or rejuvenation of hydrothermal activity. Many thermal features, such as Castle and Old Faithful geysers, have evidence of trees that are now encrusted by silica. The duration from initial tree kill to complete desiccation may be long enough to provide a useful chronometer for thermal activity. The 14C date of a small bobby sox tree near Gem Pool in the Upper Geyser Basin yielded an age of 190 years BP. The pulses and pauses documented by 14C dating of Castle Geyser, and observed in the nature of tree growth and subsequent hydrothermal kill, may be combined to develop a chronology of hydrothermal activity which, when combined with other data sets, may help provide clues to deeper processes in the Yellowstone caldera.
V53B-1335
Patterns of Stream Flow and Temperature at Tantalus Creek, Norris Geyser Basin, Yellowstone National Park (USA)
We analyzed data for stream flow and water temperature from Tantalus Creek in the Norris Geyser Basin and their relationship to air temperature, precipitation and geyser eruptions during calendar year 2005. The creek is of special interest because ~97% of its waters are derived directly from outflow of Norris Geyser Basin hot springs. Understanding and tracking observed patterns and background behavior would improve our ability to reliably detect transient episodes of anomalous hydrothermal flow at Norris. From our analysis of the data, two separate diurnal patterns emerge: 1) in winter, water temperature and stream flow closely track those of air temperature; that is, water discharge and temperature increase during the day, and decrease at night. 2) in summer, water and air temperature are closely aligned but stream flow declines as soon as water temperature reaches its daily maximum. The winter pattern is present when the average daily temperature consistently drops below 0°C whereas the summer pattern is recognizable when the average daily temperature regularly exceeds 0°C. Spring and fall systematics are much more irregular, though both summer and winter patterns can be discerned occasionally during those seasons. We interpret increases in stream flow associated with the winter pattern to result from addition of water from melted snow and ice that increases in volume once air temperature increases in the morning. Melting is facilitated by the warm ground temperatures in the geyser basin, which are significantly higher than air temperatures in the winter. The summer pattern, however, appears to be strongly affected by increased evaporation in the afternoon, which decreases flow and buffers the temperature increase. In summary, the temperature of the air and the temperature difference between air and creek water appears to drive two distinct, seasonal hydrologic patterns. Finally, we note that discharge from eruptions at Echinus Geyser are clearly visible as peaks in the hydrograph, indicating that water from this geyser reaches the weir in 80 to 90 minutes, reflecting a slug of ~33,000 liters that travels about 0.4 meters per second.